A method and system for improving reading ability based on artificial intelligence
By analyzing the text's logical structure using natural language processing and logical reasoning algorithms, a multi-level question sequence is generated for detection and feedback adjustments, optimizing reader interaction. This solves the problem of insufficient logical recognition and comprehension depth in existing reading methods and enhances readers' systematic thinking ability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-04-22
- Publication Date
- 2026-07-14
AI Technical Summary
Existing reading methods are insufficient to effectively improve readers' ability to recognize and understand the logic of texts. They lack dynamic feedback and systematic thinking training, resulting in an inability to form systematic thinking and knowledge transfer when faced with complex information.
By extracting the core elements and logical structure of the text through a natural language processing model, a multi-level question sequence is generated for real-time detection. Combined with a feedback generation module, dynamic adjustment prompts are provided to optimize the reader interaction sequence and update the system thinking model to achieve knowledge transfer.
It enhances readers' understanding of textual logic and their ability to think systematically, strengthens their thinking and adaptability when faced with complex information, and achieves a comprehensive improvement in reading ability.
Smart Images
Figure CN122389879A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information technology, and in particular to a method for improving reading ability. Background Technology
[0002] Reading ability, as a core pillar of human learning and cognitive development, plays a vital role in education, career, and daily life. Especially in this information-saturated era, efficiently acquiring and deeply understanding textual content is not only crucial for personal knowledge accumulation but also directly impacts decision-making quality and innovation capabilities. However, despite the widespread recognition of the importance of reading ability, how to scientifically and effectively improve this skill remains a key issue that urgently needs to be addressed in the fields of education and technology.
[0003] Currently, while many methods aim to improve reading comprehension through practice or tool-assisted learning, most solutions neglect the reader's active participation and dynamic feedback during the comprehension process, focusing instead on superficial memorization and retelling of content. This approach fails to grasp the underlying logic and deeper meaning of the text, often hindering readers from developing systematic thinking when faced with complex information and making it difficult to address knowledge transfer issues across different contexts. In other words, existing methods lack effective depth-of-understanding techniques and fail to truly help readers build a comprehensive understanding of the content.
[0004] A deeper challenge lies in how to capture and verify the reader's grasp of the text's logic during the reading process. This challenge is first reflected in the identification of the text's internal connections. Readers often struggle to actively discover the relationships between content, such as causal relationships or implicit assumptions, causing comprehension to remain at a fragmented level. This deficiency in identification further leads to another problem: the lack of effective means to verify whether readers have truly understood the text's deeper meaning. For example, when faced with a story, a reader may remember the plot but cannot answer questions such as what would have happened if a certain event had not occurred. This progressive dilemma from identification to verification constitutes a core technical obstacle to improving reading comprehension.
[0005] Therefore, designing a mechanism during the reading process that helps readers understand the text's internal logic while simultaneously assessing their depth of comprehension through targeted questions or tasks has become a critical issue that urgently needs to be addressed. Solving this problem is not only about improving reading comprehension but also directly related to enabling each reader to possess stronger critical thinking and adaptability skills when faced with complex information. Summary of the Invention
[0006] This invention provides a method for improving reading ability, mainly including: By extracting keywords and sentence structure from the input text using a natural language processing model, a preliminary semantic representation is obtained, which includes entity relations and dependency trees, thereby determining the core element set of the text. Based on the core element set, logical reasoning algorithms are used to analyze causal chains and hypothetical relationships to obtain the internal logical structure. The internal logical structure describes the progressive relationship and implicit conditions between elements, thereby judging the completeness of the logical connection. If the completeness of the logical connection is higher than a preset threshold, then the reader's response data is obtained by generating a targeted question sequence. The question sequence is designed as a multi-level query based on the internal logical structure to obtain the matching score of the response data. The quantitative indicators of understanding depth are determined from the matching score of the response data. The quantitative indicators integrate response consistency and logical inference accuracy to obtain the depth verification results. A feedback generation module is used to generate dynamic adjustment prompts based on the deep verification results. These prompts include strengthening the logical connections and supplementing explanations for missing parts, resulting in an optimized reader interaction sequence. The system thinking model is updated by optimizing the reader interaction sequence, in which the system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for transferred knowledge; The feasibility of knowledge transfer is determined from the knowledge transfer mapping framework. If the feasibility meets the conditions, the final reading improvement path is output, which covers the complete chain from logical connection to system thinking.
[0007] Furthermore, the natural language processing model is used to extract keywords and sentence structure from the input text to obtain a preliminary semantic representation, which includes entity relations and dependency trees, thereby determining the core element set of the text.
[0008] Furthermore, the question sequence is designed as a multi-level query based on its inherent logical structure to obtain a matching score for the response data.
[0009] Furthermore, the process involves using logical reasoning algorithms to analyze causal chains and hypothetical relationships based on the core element set, thereby obtaining an internal logical structure. This internal logical structure describes the progressive relationships and implicit conditions between elements, thus determining the completeness of the logical connections.
[0010] Furthermore, if the completeness of the logical connection is higher than a preset threshold, then reader response data is obtained by generating a targeted question sequence, wherein the question sequence is designed as a multi-level query based on the internal logical structure to obtain a matching score for the response data.
[0011] Furthermore, the quantitative index of understanding depth is determined from the matching score of the response data, wherein the quantitative index integrates response consistency and logical inference accuracy to obtain the depth verification result.
[0012] Furthermore, the feedback generation module generates dynamic adjustment prompts based on the deep verification results. These prompts include strengthening logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence.
[0013] Furthermore, the system thinking model is updated through an optimized reader interaction sequence, wherein the system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for transferred knowledge.
[0014] Furthermore, the feasibility of knowledge transfer is determined from the knowledge transfer mapping framework. If the feasibility condition is met, the final reading improvement path is output, which covers the complete chain from logical connection to system thinking.
[0015] Another object of the present invention is to provide a system for improving reading ability, comprising: The semantic extraction module is used to extract keywords and sentence structure from the input text through a natural language processing model to obtain a preliminary semantic representation, which includes entity relations and dependency trees, thereby determining the core element set of the text. The logic analysis module is used to analyze causal chains and hypothetical relationships based on the core element set using logical reasoning algorithms to obtain the internal logical structure. The internal logical structure describes the progressive relationship and implicit conditions between elements, thereby judging the integrity of the logical connection. The integrity judgment module is used to obtain reader response data by generating a targeted question sequence if the integrity of the logical connection is higher than a preset threshold. The question sequence is designed as a multi-level query based on the internal logical structure to obtain the matching score of the response data. The question generation module is used to determine the quantitative indicators of understanding depth from the matching score of the response data. The quantitative indicators integrate response consistency and logical inference accuracy to obtain the depth verification results. The deep quantification module is used to generate dynamic adjustment prompts based on the deep verification results using the feedback generation module. These dynamic adjustment prompts include strengthening paths for logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence. The feedback adjustment module is used to update the system thinking model through an optimized sequence of reader interactions. The system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for transferred knowledge. The model update and transfer judgment module is used to determine the feasibility of knowledge transfer from the knowledge transfer mapping framework. If the feasibility meets the conditions, the final reading improvement path is output, which covers the complete chain from logical connection to system thinking. Attached Figure Description
[0016] Figure 1 This is a flowchart illustrating a method for improving reading ability according to the present invention. Detailed Implementation
[0017] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.
[0018] Exemplary embodiments of this disclosure provide a method for improving reading ability. (See reference...) Figure 1 As shown, the method may include the following steps S101 to S107: Step S101: Extract keywords and sentence structure from the input text using a natural language processing model to obtain a preliminary semantic representation.
[0019] Semantic representation includes entity relations and dependency trees, thereby identifying the core set of elements in the text. In one optional implementation, the natural language processing model can employ a deep learning network to parse the input text stream. For example, when a user inputs a discussion of economic principles, the system automatically identifies core entities such as "inflation" and "monetary policy," analyzes the subject-verb-object relationships of these entities in the sentence, and constructs a dependency tree reflecting grammatical connections. In this way, the system can transform fragmented textual information into structured semantic nodes, clarifying the key objects involved in the text and their initial interactions, providing foundational data for subsequent in-depth logical analysis.
[0020] Step S102: Based on the core element set, use a logical reasoning algorithm to analyze the causal chain and hypothesis association to obtain the internal logical structure.
[0021] The internal logical structure describes the progressive relationships and implicit conditions between elements, thereby determining the completeness of logical connections. In this implementation, the logical reasoning algorithm traverses the core element set, identifies explicit logical connectors in the text, and uses semantic reasoning techniques to uncover unspoken derivation paths between paragraphs. For example, if the text mentions "due to rising raw material prices, the final product price has increased," the system will identify the causal chain and further analyze whether there are implicit assumptions such as "stable market demand." By constructing this logical network containing explicit and implicit relationships, the system can assess the rigor of the text's logic and calculate a score reflecting logical completeness to determine whether to trigger subsequent interactive verification processes.
[0022] Step S103: If the completeness of the logical connection is higher than a preset threshold, then obtain reader response data by generating a targeted question sequence.
[0023] The question sequence is designed as a multi-level query based on the text's inherent logical structure, yielding a matching score for the response data. Specifically, when the system determines that the text's logical structure is sufficiently complete, it automatically generates a series of questions, progressing from simple to complex, based on key nodes within the internal logical structure. For example, a first-level question might require the reader to identify direct causal relationships in the text, while a second-level question might require the reader to infer changes in the logical chain when a specific hypothesis changes. The reader's answers are converted into semantic vectors and compared with the system's preset logical paths. By calculating the overlap between the reader's answers and the standard logical model, the system can derive a quantitative matching score to measure the reader's initial understanding of the text's logic.
[0024] Step S104: Determine a quantitative indicator of understanding depth from the matching score of the response data.
[0025] The quantitative indicator integrates response consistency and logical inference accuracy to obtain a depth verification result. In one optional implementation, response consistency reflects whether the reader's logical thinking remains coherent when answering different logically related questions; logical inference accuracy reflects the reader's ability to capture implicit information in the text. The system performs a comprehensive weighted analysis of the matching scores generated from multiple rounds of interaction, eliminating the influence of random factors, thereby deriving a quantitative indicator that can truly reflect the reader's cognitive depth. For example, if a reader can still give an accurate and coherent answer when faced with complex logical transition questions, the system will determine that their comprehension depth is high and output the corresponding depth verification result as a basis for subsequent dynamic adjustment of teaching strategies.
[0026] Step S105 involves using a feedback generation module to generate dynamic adjustment prompts based on the deep verification results. These prompts include strengthening paths for logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence. In one possible implementation, the feedback generation module uses the comprehension depth quantification index determined in the preceding steps to precisely pinpoint the reader's weak points in the logical chain. For example, if the deep verification results show insufficient understanding of the implicit conditions of a specific causal relationship, the system will automatically generate a strengthening path for that point and provide detailed supplementary explanations. These prompts not only correct the reader's cognitive biases but also construct a more targeted optimized interaction sequence by adjusting the difficulty and focus of subsequent questions, guiding the reader to gradually improve their logical understanding.
[0027] Step S106 involves updating the system thinking model through an optimized reader interaction sequence. The system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for transferred knowledge. In this embodiment, the system thinking model not only records the reader's current cognitive state but also simulates the evolution of the reader's thinking patterns by integrating dynamic data from the feedback process. By correlating quantitative understanding depth indicators with dynamic adjustment prompts, the system can construct a cross-domain knowledge transfer mapping framework. This framework can identify the thinking characteristics exhibited by the reader when processing the current text logic and transform them into transferable cognitive strategies, providing readers with a universal logical analysis template when facing different types of complex texts.
[0028] Step S107: Determine the feasibility of knowledge transfer within the knowledge transfer mapping framework and output the final reading improvement path. If the feasibility meets preset conditions, the output will cover the complete chain from logical connections to systemic thinking. During implementation, the system will evaluate the universality and accuracy of the cognitive strategies extracted from the mapping framework. If the strategy is determined to effectively assist readers in logical deduction in other contexts, the system will generate a detailed reading improvement plan. This plan not only includes an in-depth summary of the logical connections in the current text but also extends to higher-level systemic thinking training. Through structured guidance, it helps readers establish a complete thinking chain from local logical analysis to global systemic cognition, achieving a substantial leap in reading ability.
[0029] Exemplary embodiments of this disclosure also provide a system for improving reading ability. This system mainly includes a semantic extraction module, used to extract keywords and sentence structures from input text using a natural language processing model, obtaining a preliminary semantic representation containing entity relations and dependency trees, thereby determining the core element set of the text. A logical analysis module is then used to analyze causal chains and hypothetical associations based on the core element set using logical reasoning algorithms, obtaining the inherent logical structure describing the progressive relationships and implicit conditions between elements, and thus determining the completeness of the logical connections.
[0030] In addition, the system includes a completeness judgment module, which, when the completeness of logical connections exceeds a preset threshold, generates a multi-level targeted question sequence based on the inherent logical structure to obtain reader response data and calculate a matching score. The question generation module is responsible for determining a comprehension depth quantification index that integrates response consistency and logical inference accuracy from the matching score of the response data, thereby obtaining a deep verification result. The depth quantification module, based on the deep verification result, uses the feedback generation module to generate dynamically adjusted prompts containing reinforcement paths and supplementary explanations to obtain an optimized reader interaction sequence.
[0031] Finally, the feedback adjustment module updates the system thinking model through an optimized reader interaction sequence, integrating quantitative indicators and adjustment prompts to construct a mapping framework for knowledge transfer. The model update module is responsible for determining the feasibility of knowledge transfer from this mapping framework and, when conditions are met, outputting a final reading improvement path covering the logical connections to the complete chain of system thinking. It should be noted that this application's technical solution clearly informs users of the processing rules and obtains their consent before processing personal information. For sensitive personal information, explicit consent requirements will be ensured through prominent labeling, pop-up authorization, etc., and the processor's information, processing purpose, and method will be clearly communicated to fully protect users' right to know and authorization.
[0032] If the technical solution of this application involves personal information, the product using this solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If sensitive personal information is involved, the user's separate consent has been obtained before processing, and the "express consent" requirement is met. For example, a clear sign is placed at the collection device such as a camera to inform the user that they have entered the collection area, and the user's voluntary entry is considered as consent; or the processing device clearly indicates the processing rules and obtains authorization through pop-up windows or by asking the user to upload information themselves. The personal information processing rules include the processor, the purpose of processing, the processing method, and the types of personal information.
[0033] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.
Claims
1. A method for improving reading ability based on artificial intelligence, characterized in that, The method includes: Keywords and sentence structure are extracted from the input text using a natural language processing model to obtain a preliminary semantic representation, which includes entity relations and dependency trees, thus determining the core element set of the text. Based on this core element set, a logical reasoning algorithm is used to analyze causal chains and hypothesis associations to obtain the internal logical structure. This internal logical structure describes the progressive relationships and implicit conditions between elements, thereby judging the completeness of logical connections. If the completeness of the logical connections exceeds a preset threshold, a targeted question sequence is generated to obtain reader response data. This question sequence is designed as a multi-level query based on the internal logical structure, resulting in a matching score for the response data. A quantitative indicator of understanding depth is determined from the matching score of the response data. This quantitative indicator integrates response consistency and logical inference accuracy, resulting in a depth verification result. A feedback generation module generates dynamic adjustment prompts based on the depth verification result. These prompts include strengthening paths for logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence. The system thinking model is updated by optimizing the reader interaction sequence. The system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for knowledge transfer. The feasibility of knowledge transfer is judged from the mapping framework. If the feasibility meets the conditions, the final reading improvement path is output. The reading improvement path covers the complete chain of logical connection to system thinking.
2. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The process involves extracting keywords and sentence structure from the input text using a natural language processing model to obtain a preliminary semantic representation, which includes entity relations and dependency trees, thereby determining the core element set of the text.
3. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The process involves using logical reasoning algorithms to analyze causal chains and hypothetical relationships based on the core element set, thereby obtaining the internal logical structure. This internal logical structure describes the progressive relationships and implicit conditions between elements, thus determining the completeness of the logical connections.
4. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, If the completeness of the logical connection is higher than a preset threshold, then the reader's response data is obtained by generating a targeted question sequence. The question sequence is designed as a multi-level query based on the internal logical structure to obtain the matching score of the response data.
5. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The quantitative index of understanding depth is determined from the matching score of the response data. The quantitative index integrates response consistency and logical inference accuracy to obtain the depth verification result.
6. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The feedback generation module generates dynamic adjustment prompts based on the deep verification results. These prompts include strengthening logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence.
7. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The system thinking model is updated through an optimized reader interaction sequence, wherein the system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for transferred knowledge.
8. The method for improving reading ability based on artificial intelligence according to claim 1, characterized in that, The feasibility of knowledge transfer is determined from the knowledge transfer mapping framework. If the feasibility meets the conditions, the final reading improvement path is output, which covers the complete chain from logical connection to system thinking.
9. A system for improving reading ability, characterized in that, The system includes: a semantic extraction module, used to extract keywords and sentence structure from input text using a natural language processing model to obtain a preliminary semantic representation, which includes entity relations and dependency trees, thereby determining the core element set of the text; a logical analysis module, used to analyze causal chains and hypothesis associations based on the core element set using logical reasoning algorithms to obtain the internal logical structure, which describes the progressive relationships and implicit conditions between elements, thereby judging the completeness of logical connections; a completeness judgment module, used to obtain reader response data by generating a targeted question sequence if the completeness of the logical connection is higher than a preset threshold, where the question sequence is designed as a multi-level query based on the internal logical structure, and obtains a matching score for the response data; a question generation module, used to determine a quantitative indicator of understanding depth from the matching score of the response data, where the quantitative indicator integrates response consistency and logical inference accuracy to obtain a depth verification result; and a depth quantification module, used to generate dynamic adjustment prompts based on the depth verification result using a feedback generation module, where the dynamic adjustment prompts include strengthening paths for logical connections and supplementary explanations for missing parts, resulting in an optimized reader interaction sequence. The feedback adjustment module is used to update the system thinking model through an optimized reader interaction sequence. The system thinking model integrates quantitative indicators and adjustment prompts to obtain a mapping framework for knowledge transfer. The model update and transfer judgment module is used to judge the feasibility of knowledge transfer from the mapping framework for knowledge transfer. If the feasibility meets the conditions, the final reading improvement path is output, which covers the complete chain of logical connections to system thinking.